The modulation of RNA structure is an essential regulatory process in many cellular events, such as, for example, pre-mRNA splicing, assembly of spliceosomes, assembly of ribosomes, protein translation, which can be summarized under the generic term "regulation of gene expression at the RNA level". The so-called "DEAD box" protein family of putative RNA helicases, named after the characteristic amino acid motif Asp-Glu-Ala-Asp (in the single-letter code DEAD) (SEQ ID NO:35), in this context plays a key part (in particular for the modulation of the secondary and tertiary structure of mRNA). DEAD box proteins are also involved in processing of DNA. The members of this family and some subfamilies have differences in their specific function and cellular localization. However, in addition to characteristic sequence homologies certain members also show similar biochemical properties (F. V. Fuller-Pace, Trends in Cell Biology, Vol 4, 1994, 271-274). The characteristic protein sequences of the DEAD proteins are highly conserved in evolution (S. R. Schmid and P. Lindner, Molecular and Cellular Biology, Vol 11, 1991, 3463-3471). Members of this protein family are found in various viruses, bacteria, yeasts, insects, molluscs and lower vertebrates up to mammals and are responsible for a large number of cellular functions. The fact that even relatively simple organisms such as, for example, the yeast Saccharomyces cerevisiae express numerous proteins of the DEAD box protein family and their subfamilies, suggests that each of these proteins contributes to the specific interaction with certain RNAs or RNA families (I. lost and M. Dreyfus, Nature Vol 372, 1994, 193-196). It has been shown that translation factors, such as elF-4A and the proteins involved in the pre-mRNA splicing process, recognize specific RNA target sequences or structures. Nevertheless, to date there is little information about the structure and the synthesis of characteristic RNA sequences which require the DEAD proteins for recognition and for ATPase/RNA helicase reaction (A. Pause and N. Sonenberg, Current Opinion in Structural Biology Vol 3, 1993, 953-959).
The DEAD box protein family is an enzyme class which is growing and which is involved in the various reactions in post transcriptional regulation of gene expression. Because of the high number of different cellular DEAD box proteins, it is to be expected that specific RNA helicases are assigned to certain classes of gene products, e.g. viral proteins, heat shock proteins, antibody and MHC proteins, receptors, RNAs etc. This specificity indicates that members of this protein family are attractive pharmacological targets for active compound development.
Two of the subclasses of the DEAD (SEQ ID NO:35) box protein family are the DEAH (SEQ ID NO:36) proteins (having one specific amino acid replacement) and the DEXH (SEQ ID NO:37) protein (having two amino acid replacements in the main motif, X being any desired amino acid) families, which also play a part in the replication, recombination, repair and expression of DNA and RNA genomes (Gorbalenya, A. E., Koonin, E. V., Dochenko, A. P., Blinov, V. M., 1989: Nucleic Acids Res. 17, 4713-4729). The DEAD box proteins and their subfamilies are often designated "helicase superfamily II " (Koonin, E. V., Gorbalenya, A. E., 1992: FEBS 298, 6-8). This superfamily has seven highly conserved regions. Altogether, up to now over 70 members belong to this superfamily II.
The following schematic representation of the DEAD family and the DEAH and DEXH families subfamilies (Schmid, S. R., Lindner P., 1991: Molecular and Cellular Biology 11, 3463-3471) shows the similarity between the families. The structure of elF-4A, a member of a DEAD box protein, is also shown. The numbers between these regions show the distances in amino acids (AA). X is any desired AA. Where known, functions have been assigned to the ranges.
DEAD (SEQ ID NO:35) FAMILY ATPase A motif ATPase B motif NH.sub.2 -----AXXXGKT----PTRELA-----GG---TPGR---DEAD--- SAT---FXXXT--- 21-299 24-42 22-28 19-27 19-22 27-51 59-70 52-53 RGXD-----HRIGRXXR---COOH (SEQ ID NO:19) 20 24-236 elF-4A NH.sub.2 -----AXXXXGKT---PTRELA-----GG---TPGR---DEAD--- SAT---FINT--- 75 24 22 20 20 27 62 52 RGID------HRIGRXXR---COOH (SEQ ID NO:20) 20 41 DEAH (SEQ ID NO:36) SUBFAMILY NH.sub.2 -----GXXXXGKT---RVAA-----XX---TDGX---DEAH---SA T---FXT--- 245-505 22-24 29 7-8 19 28 58-61 75-84 XGXX------QRIGRXGR---COOH (SEQ ID NO:21) 25 315-373 DEXH (SEQ-ID NO:37) SUBFAMILY NH.sub.2 -----XXXXXGKT---PTRXXX----------------DEXH---T AT---FXXS--- 81-1904 19-27 55-60 24-30 44-72 46-55 XGXX------QRXGRXGR---COOH (SEQ ID NO:22) 38-44 155-1799
The ATPase motif (AXXXXGKT, SEQ ID NO:23) is an amino-terminal conserved region and occurs in most proteins which bind nucleotides, i.e. also in other proteins which interact with DNA and RNA, such as DNAB (part of the primosome), UvrD (endonuclease), elongation factor 1 and transcription termination factor Rho (Ford M. J., Anton, I. A., Lane, D. P., 1988: Nature 332, 736-738). As used in this specification "ATPase activity" is used to mean the ability to catalyze hydrolysis of ATP. The ATPase A and ATPase B motifs function together in the enzymatic process of ATP hydrolysis.
The second conserved region is the so-called DEAD (SEQ ID NO:35) box, or DEAH (SEQ ID NO:36), DEXH (SEQ ID NO:37) or DEXX (SEQ ID NO:38) box in other families of the helicases and nucleic acid-dependent ATPases. This region represents the ATPase B motif. In the reaction mechanism, the N-terminal aspartic acid in the DEAD box binds Mg.sup.2+ via a water molecule (Pai, E. F., Krengel, U., Petsko, G. A., Gody, R. S., Katsch, W., Wittinghofer, A., 1990: EMBO J. 9, 2351-2359). Mg.sup.2+ in turn forms a complex with the .beta.- and gamma-phosphate of the nucleotide and is essential for the ATPase activity. Substitutions of the first two amino acids of the DEAD region in elF-4A prevent ATP hydrolysis and RNA helicase activity, but not ATP binding (Pause, A., Sonenberg, N., 1992: EMBO J. 11, 2643-2654). The DEAD region additionally couples RNA helicase activity to ATPase activity. The hydrolysis of ATP provides the energy needed for RNA unwinding during helicase activity.
The third region investigated is the SAT region (sometimes also TAT). As a result of mutation in this region, RNA helicase activity is suppressed, but other biochemical properties are retained (Pause A. & Sonenberg N., 1992). As used in this specification "helicase activity" is used to mean the ability to directly or indirectly catalyze the unwinding of RNA.
The farthest carboxy-terminal region is the HRIGRXXR (SEQ ID NO:24) region, which is necessary for RNA binding and ATP hydrolysis.
As stated above, members of the DEAD box protein family bind ATP and nucleic acid. As used in this specification a protein that "binds nucleic acid" is defined as a protein that forms complexes with nucleic acid. The binding can be measured by standard methods like Electrophoretic Mobility Shift Assay (EMSA) or ELISA, which are well known in the art. The following assays may also be used: Scintillation Proximity Assay (SPA, Amersham International, Little Chalfont, Buckinghamshire, England) and BlAcore (Biomolecule Interaction Analysis, Pharmacia, Upsala Sweden).
As used in this specification, a protein that "binds ATP" is defined as a protein that will bind ATP as measured using an assay that measures ability of labeled ATP to bind to protein. The ATP may be labeled using radioactive or fluorescent label. One example of an ATP binding assay is described in Pause, et al. EMBO J. 11:2643 (1992), which is hereby incorporated by reference. Briefly, a protein according to the invention is incubated in a crosslinking reaction mixture containing Tris-HCl (pH 7.5), Mg acetate, .sup.32 P-ATP, glycerol and DTT in the presence or absence of poly(u) (Pharmacia) under a 15 watt germicidal lamp at 4.degree. C. Unlabelled ATP is then added, followed by addition of RNase A at 37.degree. C. Samples are boiled in SDS-PAGE sample buffer and electrophoresed.
It follows from the above-mentioned relationships that specific RNA helicases are attractive targets for pharmaceutically active substances. For example, it is also known that certain pathogenic viruses, which can cause diseases in humans, animals or plants, carry in their genome a gene encoding an RNA helicase, which is needed for accurate replication (E. V. Koonin, 1991). Thus, specific substances that interfere with, or modulate, the activity of such virus-specific helicases could be used to treat virally-mediated diseases. Because helicases are also found in plants, substances that modulate plant helicases may be used to protect plants against pathogenic viruses. (F. V. Fuller-Pace, Trends in Cell Biology, Vol. 4, 1994, 271-274). Helicases also make attractive targets for development of therapeutic treatments for various types of diseases. For example, hereditary diseases such as Werner's syndrome and Bloom's syndrome have been linked to the production of proteins with helicase structure. See Yu, et al. Science 272: 258 (1996) and Research News, Science 272: 193 (1 996)(Werner's); Ellis, et al. Cell 83:655 (1995), and D. Bassett "Genes of Medical Interest" In http://www.ncbi.nih.gov/xREFdb/(Bloom's). A nucleolar RNA helicase is recognized by the autoimmune antibodies from a patient with watermelon stomach. Valdez, et al., Nucl. Acid. Res., 24:1220 (1996). In retinoblastoma cancer cells, expression of a DEAD box protein is amplified. Godbout, et al. Proc. Natl. Acad. Sci. USA 90:7578 (1993). In addition, RNA processing plays a role in a number of processes that are implicated in other disease states. For example, in diabetic mice, the leptin receptor is abnormally spliced. Lee, et al. Nature 379:632 (1996). In addition, post-transcriptional regulation of human interleukin-2 gene expression occurs at the level of processing of precursor transcripts, which may be linked to the presence of a protein. Gerez, et al. J. Biol Chem. 270:19569 (1995).
Thus, therapeutic agents can be designed that interfere with helicase activity or RNA processing that is associated with the disease state.
The isoxazole derivative leflunomide shows anti-inflammatory and immunosuppressive properties without causing damage to the existing functions of the immune system (HWA486 (leflunomide); R. R. Bartlett, G. Campion, P. Musikic, T. Zielinski, H. U. Schorlemmer In: A. L. Lewis and D. E. Furst (editors), Nonsteroidal Anti-inflammatory Drugs, Mechanisms and Clinical Uses (Dekker: New York, 1994); C. C. A. Kuchle, G. H. Thoenes, K. H. Langer, H. U. Schorlemmer, R. R. Bartlett, R. Schleyerbach, Transplant Proc. 1991, 23:1083-6; T. Zielinski, H. J. Muller, R. R. Bartlett, Agents Action 1993, 38: C80-2). Many activities, such as the modification of cell activation, proliferation, differentiation and cell cooperation, which can be observed in autoimmune diseases, are modulated by leflunomide or its active metabolite, A77 1726. ##STR1## Studies on the molecular mechanism of action of this active compound point to an influence on the pyrimidine metabolism. Because leflunomide is very rapidly converted in the body into A77 1726, in this specification, leflunomide and A77 1726 are used interchangeably. Thus, both "leflunomide resistance" and "A77 1726 resistance" are used to designate the same condition.
Pyrimidine and purine nucleotides play a key part in biological processes. As structural units of DNA and RNA, they are thus carriers of genetic information. The biosynthesis of the pyrimidines comprises the oxidation of dihydroorotate to orotate, which is catalyzed by the enzyme dihydroorotate dehydrogenase (DHODH). Altogether, six enzymes are needed for the de novo synthesis of uridine monophosphate (UMP). UMP plays a key part in the synthesis of the other pyrimidines, cytidine and thymidine. The inhibition of DHODH thus leads to an inhibition of pyrimidine de novo synthesis. Particularly affected are immune cells, which have a very high need for nucleotides, but can only cover a little of this by side routes (salvage pathway). Binding studies with radiolabeled leflunomide analogs identified the enzyme DHODH as a possible site of action of A77 1726 and thus the inhibition of DHODH by leflunomide is an important starting point for the elucidation of the observed immunomodulating activities. Williamson, et al. J. Biol. Chem. 270:22467-22472 (1995).